The fact that you had to choose so ridiculous an example suggests that Paul Graham is basically correct. (I think the correct reading of “anything you do to decrease one will probably also decrease the other” is “if you pick something that decreases one, it will probably decrease the other” rather than “literally every single thing that might decrease one will, with high probability given that you do that particular thing, decrease the other”.)
The fact that you had to choose so ridiculous an example suggests that Paul Graham is basically correct.
No it doesn’t. It suggests that when selecting examples for the purpose of countering generalizations wedrifid chooses examples that are clear and unambiguous to anyone who correctly parses the claim rather than choosing the most likely counter example. This is particularly the case when rejecting the extent of a general claim while accepting the gist—as I went out of the way to make explicit.
I also reject the idea that the second example I gave is at all unrealistic:
Punish all observed incidents of stupidity with physical beating.
Corporal punishment for stupidity is an actual (hopefully mostly historical) thing.
For the record, I took you to be proposing a single counterexample with two components, rather than two separate counterexamples; I’m sorry for the misunderstanding.
Now that I know the second bullet point was meant to be a separate counterexample, I have a different objection to it: I am unconvinced that any implementable version of it would both reduce stupidity and increase meanness. (The most likely outcome, I think, would be to increase meanness while replacing more blatant varieties of stupidity with more widely spread lower-level stupidity.)
EDITED to add: Oh, one other thing. If it happens that (1) it was you who downvoted me and (2) you did so because you thought I downvoted your previous comment, then you might want to know that I didn’t.
The fact that you had to choose so ridiculous an example suggests that Paul Graham is basically correct. (I think the correct reading of “anything you do to decrease one will probably also decrease the other” is “if you pick something that decreases one, it will probably decrease the other” rather than “literally every single thing that might decrease one will, with high probability given that you do that particular thing, decrease the other”.)
No it doesn’t. It suggests that when selecting examples for the purpose of countering generalizations wedrifid chooses examples that are clear and unambiguous to anyone who correctly parses the claim rather than choosing the most likely counter example. This is particularly the case when rejecting the extent of a general claim while accepting the gist—as I went out of the way to make explicit.
I also reject the idea that the second example I gave is at all unrealistic:
Corporal punishment for stupidity is an actual (hopefully mostly historical) thing.
I can’t help this quote:
--N-Space, Larry Niven
For the record, I took you to be proposing a single counterexample with two components, rather than two separate counterexamples; I’m sorry for the misunderstanding.
Now that I know the second bullet point was meant to be a separate counterexample, I have a different objection to it: I am unconvinced that any implementable version of it would both reduce stupidity and increase meanness. (The most likely outcome, I think, would be to increase meanness while replacing more blatant varieties of stupidity with more widely spread lower-level stupidity.)
EDITED to add: Oh, one other thing. If it happens that (1) it was you who downvoted me and (2) you did so because you thought I downvoted your previous comment, then you might want to know that I didn’t.